Abstract
Background
Exertional breathlessness is a common and disabling symptom for people with COPD and interstitial lung disease (ILD). Fan-to-face (F2F) therapy, which may stimulate trigeminal nerve afferents via cool facial airflow, is a promising intervention for breathlessness; however, its efficacy during standardised exercise is not well established. The objective was to investigate the acute effect of F2F therapy versus fan-to-leg (F2L) and no fan (NF) control conditions on breathlessness during a 4-min constant-work rate treadmill exercise test (4-min CWR-TT) among people with COPD or ILD.
Methods
In a randomised crossover trial, 32 participants (COPD: n=27; ILD: n=5) completed three 4-min CWR-TTs with F2F, F2L or NF. Breathlessness intensity (primary outcome) and physiological variables were recorded throughout.
Results
F2F reduced breathlessness intensity by 0.5 units on the Borg 0–10 scale, compared with NF, but did not reduce breathlessness intensity compared with F2L. F2F resulted in a lower facial skin temperature, but no differences were observed in ventilation, breathing pattern, heart rate or blood oxygen saturation. In an exploratory subgroup analysis of 20 participants with breathlessness intensity of ≥3 Borg units under NF, F2F decreased breathlessness intensity by 0.9 units during the 4-min CWR-TT, compared with NF, but not compared with F2L. Clinically meaningful reductions in exertional breathlessness intensity (≥1 Borg unit) were reported by 31% of the full sample and 45% of the exploratory subgroup under F2F versus NF, with similar proportions (31% and 40%) observed under F2L versus NF.
Conclusion
Whether due to stimulation of trigeminal nerve afferents or placebo effects, F2F therapy should reasonably be considered as a potentially effective nonpharmacological intervention for exertional breathlessness, especially among adults with chronic lung disease who report more severe breathlessness on exertion.
Shareable abstract
Fan-to-face (F2F) therapy can reduce exertional breathlessness in adults with chronic lung disease independently of changes in physiological variables. F2F therapy represents an affordable, feasible, nonpharmacological intervention for breathlessness. https://bit.ly/46rOy8z
Introduction
Despite optimal medical management, breathlessness (dyspnoea [1]) remains a prevalent and disabling symptom for people with COPD [2] or interstitial lung disease (ILD) [3]. Exertional breathlessness is a major contributor to exercise intolerance among those with COPD or ILD. Effective, accessible and affordable interventions are urgently needed.
Blowing air onto the face – fan-to-face (F2F) therapy – is a promising adjuvant strategy for breathlessness relief [4–9]. Potential mechanisms involved include stimulation of the transient receptor potential melastatin-8 (TRPM8) channels on the trigeminal nerve, potentially modulating activity in brain regions involved in breathlessness perception (e.g., insular cortex, amygdala, thalamus), with or without reductions in inspiratory neural drive [10, 11].
As summarised in systematic reviews and empirical studies in chronic lung disease and cancer patients [5, 7, 12–15], most F2F studies show reduced breathlessness at rest or during recovery from exercise. Only three studies have explored its effect during exercise [4, 15, 16]. Marchetti et al. [4] found that people with severe COPD exercised 34% longer and reported less breathlessness (by 1.5 units on the modified Borg 0–10 Category-Ratio scale (mBorg-CR10)) during constant-work rate (CWR) cycling with F2F versus fan to leg (F2L). Long et al. [16] showed that symptomatic adults with COPD reported lower levels of exercise-induced breathlessness (by 1 unit on a 0–10 numeric rating scale) and walked a greater distance (by 28 m) when they performed the 6-min walk test (6MWT) with a handheld fan blowing air onto their face than without a fan. Venkatapen et al. [15] also found lower levels of end-exercise breathlessness (by 1 unit on the mBorg-CR10 scale) and greater distance walked at the end of a 6MWT with a portable fan around the neck than with no fan. Across all studies, physiological variables such as breathing pattern and ventilation were largely unchanged, supporting a central mechanism of facial airflow [10, 11, 17]. However, several limitations remain. Marchetti et al. [4] presumably used a face mask (or mouthpiece and nose clip) during cardiopulmonary exercise testing (CPET) that may have blocked airflow to intranasal TRPM8 receptors. The 6MWT used by Long et al. [16] and Venkatapen et al. [15] is not a standardised exertion test, which undermines its utility to assess for treatment effects on exertional breathlessness [15]. All three studies had relatively small samples (10–25 people with chronic lung disease), and used different control conditions: F2L [4] or no fan (NF) [15, 16]. Thus, an adequately powered study assessing the effects of F2F therapy during a standardised exercise protocol, using both active (F2L) and inactive (NF) controls without obstruction of nasal airflow, is warranted.
The primary aim of this study was to investigate the acute effect of F2F versus NF on intensity ratings of exertional breathlessness during a 4-min CWR treadmill exercise test (4-min CWR-TT) in adults with COPD or ILD. Secondary aims were to investigate the acute effects of F2F versus F2L on exertional breathlessness, and to examine physiological responses to exercise across all conditions. We hypothesised that, compared with NF, F2F therapy would decrease intensity ratings of exertional breathlessness, without a corresponding change in the physiological response to exercise.
Methods
This randomised controlled crossover trial received ethics approval from the Research Institute of the McGill University Health Centre (IRB# 2023-9286) and was registered on ClinicalTrials.gov (NCT05659550). After providing written and informed consent, men or women aged ≥40 years with COPD or ILD completed three study visits. Detailed inclusion and exclusion criteria can be found in the supplemental material. Visit 1 consisted of an evaluation of participant-reported breathlessness [18, 19] and health status [20, 21], post-bronchodilator (400 μg of salbutamol) pulmonary function testing and symptom-limited incremental CPET on a motorised treadmill (Trackmaster 428CP Medical Treadmill, Full Vision, Inc., Newton, KS, USA) to determine peak power output (PPO) and peak oxygen consumption (V′O2peak). Visit 2 comprised symptom-limited CWR CPET on a motorised treadmill at 75% of PPO. Visit 3 consisted of three 4-min CWR-TTs under a F2F, F2L or NF condition, randomised to order. Additional details can be found in the supplemental material.
Cardiopulmonary exercise testing
Following a baseline (pre-exercise) rest period of ≥6 min, symptom-limited incremental CPET and CWR CPET consisted of increases in power of 10 W every 2 min and a step increase in power output to 75% PPO, respectively. Additional details can be found in the supplemental material.
4-min constant-work rate treadmill test
The 4-min CWR-TT was used for standardised assessment of exertional breathlessness at a steady-state intensity, while minimising fatigue and maximising participant tolerability. Each 4-min CWR-TT was separated by a rest period of 15 min to allow for adequate recovery time. The 4-min CWR-TT consisted of a 3-min pre-exercise baseline period followed by a step increase in exercise intensity. The exercise intensity (power output) selected for each participant's 4-min CWR-TT was based on their exercise endurance time during the CWR CPET. If a participant's exercise endurance time was <3 min, 3–8 min or >8 min, they performed each 4-min CWR-TT at 50%, 75% or 90% PPO, respectively. Refer to supplemental figure 1 for additional details. At the end of the pre-exercise baseline period and within the last 15 s of each 1-min interval during the 4-min CWR-TT, participants rated the intensity and unpleasantness of their perceived breathlessness and the intensity of their perceived leg discomfort using the mBorg-CR10 scale [22, 23]. The end points of the mBorg-CR10 scale were anchored prior to each 4-min CWR-TT such that 0 represented “no breathlessness intensity, breathlessness unpleasantness or leg discomfort” and 10 represented “the most severe breathlessness intensity, breathlessness unpleasantness or leg discomfort you have ever experienced or can imagine experiencing”. Researchers used standardised scripts to assess exertional symptoms and to help participants distinguish between breathlessness intensity and unpleasantness (see supplemental material for more details).
Prior to the first 4-min CWR-TT, participants were fitted with a Hexoskin Smart Shirt™ (Carré Technologies, Inc., Montreal, QC, Canada) individualised to their body size and gender according to the manufacturer's recommendations for proper recording of physiological variables, including respiratory rate (RR), tidal volume (VT), minute ventilation (V′E) and heart rate (HR). Peripheral oxygen saturation (SpO2) was monitored by finger pulse oximetry. The Hexoskin Smart Shirt™ was used instead of a respired flow sensor with face mask (or mouthpiece and nose clip) because it allowed for measurement of RR, VT and V′E without obstructing airflow to the nose and mouth (additional details can be found in the supplemental material). This helped to ensure that airflow applied to the face during F2F therapy could reach and stimulate intranasal/upper-airway TRPM8 channels on the trigeminal nerve.
Interventions: fan to face and fan to leg
A Honeywell HT-908 TurboForce Room Air Circulator (60 W, 120 V, 3×12″ blades, Honeywell International Inc., Charlotte, NC, USA) was mounted to an intravenous drip pole with a C-clamp so that the position of the fan could be adjusted to direct airflow to the second (forehead) and third (cheek) branches of the trigeminal nerve in the face or the side of the lower leg (F2L) [10, 11]. As no standardised fan speed for breathlessness relief currently exists in the literature, participants self-selected fan speed for optimal facial airflow comfort and temperature. The corresponding airflow (speed) was measured at the face with a digital anemometer. Researchers did not reference optimisation of breathlessness when deciding on optimal fan speed to prevent bias; only general comfort was mentioned during the study period. All participants were fan naïve as well as being naïve to the study hypothesis. During F2L, the fan was positioned so that the same airflow as F2F was directed at the calf of the participant's exposed lower left leg. F2L was chosen as the active control to help separate the effects of general airflow from the specific effects of F2F and to account for a possible placebo effect of F2F on exertional breathlessness. Using a medical-grade infrared contactless digital thermometer, forehead and cheek skin surface temperature were recorded at the end of the pre-exercise baseline period and within the last 15 s of each 1-min interval during the 4-min CWR-TT. Facial skin surface temperature (Tface) was calculated as the average of forehead and cheek temperatures and used for analysis.
Outcome variables
The primary outcome was mBorg-CR10 scale intensity ratings of breathlessness at the end of the 4-min CWR-TT between F2F and NF conditions. Secondary outcomes were mBorg-CR10 scale intensity ratings of breathlessness at the end of the 4-min CWR-TT between F2F and F2L and between F2L and NF conditions, and measures of Tface, HR, SpO2, RR, VT, V′E and ratings of breathlessness unpleasantness and leg discomfort at the end of the 4-min CWR-TT across all conditions.
Sample size
Using a repeated-measures, within-factor ANOVA framework (α=0.05, power=0.80), assuming a correlation of 0.50 among repeated measures, we estimated that a minimum of 21 participants would be required to detect a minimal clinically important difference of ±1 unit on the mBorg-CR10 scale [24] in exertional breathlessness intensity, with a standard deviation of 1.6 units at the end of the 4-min CWR-TT between F2F and NF conditions. The standard deviation and corresponding standardised effect size (Cohen's f=0.31) were derived from published [25, 26] and unpublished intervention trial (Dennis Jensen, McGill University, Montreal, QC, Canada; personal communication) data in COPD from our laboratory.
Statistical analyses
Primary and secondary outcomes were analysed using a one-way repeated-measures ANOVA to assess differences across experimental conditions (F2F, F2L, NF). When a significant main effect was observed, pairwise comparisons were performed using Tukey's test to control for multiple comparisons. Analyses were performed with SigmaStat (version 3.5, Systat Software, San Diego, CA, USA). Statistical significance was set at p<0.05. Values are reported as mean±sd unless otherwise stated.
To reduce the risk of a floor effect masking the potential benefits of F2F therapy on exertional breathlessness, we conducted an exploratory unplanned subgroup analysis on 20 participants with breathlessness intensity ratings of ≥3 units on the mBorg-CR10 scale at the end of the 4-min CWR-TT under the NF condition.
Results
A total of 36 participants were assessed for eligibility and enrolled in the study. However, two participants with COPD were excluded due to exacerbations, one with COPD was lost to follow-up after suffering a musculoskeletal injury unrelated to the study procedures and one with COPD was lost to follow-up due to transportation issues getting to the testing location (supplemental figure 2). 32 participants (14 females) completed all three visits; their baseline characteristics are presented in table 1. Participants were symptomatic adults with COPD (n=27, 84%) or ILD (n=5, 16%) characterised by impaired lung function and exercise intolerance (mean V′O2peak<15 ml·kg−1·min−1 and <80% predicted), mostly due to intolerable breathlessness.
TABLE 1.
Baseline participant characteristics (n=32)
| Parameter | Value |
|---|---|
| Female, n (%) | 14 (44) |
| Age, years | 69.5±8.1 |
| Body mass index, kg·m−2 | 27.5±9.4 |
| Clinical diagnosis, n (%) | |
| COPD | 27 (84) |
| GOLD 1 | 3 (9) |
| GOLD 2 | 5 (16) |
| GOLD 3 | 14 (44) |
| GOLD 4 | 5 (16) |
| Interstitial lung disease | 5 (16) |
| Idiopathic pulmonary fibrosis | 3 (60) |
| Sarcoidosis | 1 (20) |
| Undefined | 1 (20) |
| Medication summary, n (%) | |
| SABA | 22 (69) |
| SAMA | 1 (3) |
| SABA+SAMA | 1 (3) |
| LABA | 4 (13) |
| LAMA | 8 (25) |
| LABA+LAMA | 17 (53) |
| ICS | 8 (25) |
| LABA+ICS | 14 (44) |
| LABA+LAMA+ICS | 4 (13) |
| Antifibrotic | 1 (3) |
| Domiciliary oxygen | 2 (6) |
| Breathlessness and health status | |
| mMRC Dyspnoea Scale score, 0–4 | 2.2±1.1 |
| mMRC score ≥2, n (%) | 21 (66) |
| CAT score, 0–40# | 19.2±9.5 |
| CAT score ≥10, n (%) | 16 (80) |
| Baseline Dyspnoea Index, 0–12 | 5.7±2.5 |
| HADS-Anxiety score, 0–21 | 6.9±3.6 |
| HADS-Depression score, 0–21 | 5.1±3.5 |
| Post-bronchodilator lung function (% predicted) | |
| FEV1/FVC, % | 47.7±18.9 |
| FEV1, L | 1.2±0.7 (52±25) |
| Total lung capacity, L | 6.0±1.6 (105±24) |
| Inspiratory capacity, L | 1.9±0.7 (75±20) |
| Residual volume, L | 3.4±1.3 (164±58) |
| Functional residual capacity, L | 4.1±1.4 (128±44) |
| DLCO, mL·min−1·mmHg−1 | 13.0±5.1 (61±23) |
| R5, kPa·L−1·s−1 | 5.6±2.9 (188±121) |
| R20, kPa·L−1·s−1 | 3.6±1.4 (107±52) |
| R5−R20, kPa·L−1·s−1 | 2.1±1.9 (197±200) |
| X5, kPa·s−1·L−1 | −3.7±4.9 (294±432) |
| Symptom-limited peak incremental cardiopulmonary exercise test results (visit 1) | |
| Power output, W | 38.3±24.1 |
| V′O2peak, mL·kg−1·min−1 (% predicted) | 13.1±5.6 (76±73) |
| Reason(s) for stopping exercise, n (%) | |
| Breathlessness | 17 (53) |
| Leg discomfort | 7 (22) |
| Breathlessness+leg discomfort | 4 (13) |
| Other | 5 (16) |
| Symptom-limited peak constant-work rate cardiopulmonary exercise test results (visit 2) | |
| Power output, W | 29.3±18.0 |
| V′O2peak, mL·kg−1·min−1 (% V′O2peak from visit 1) | 12.5±5.4 (90±27) |
| Exercise endurance time, n (%) | 6.1±4.2 |
| <3 min | 7 (22) |
| 3–8 min | 16 (50) |
| >8 min | 9 (28) |
| Reason(s) for stopping exercise, n (%) | |
| Breathlessness | 21 (66) |
| Leg discomfort | 6 (19) |
| Breathlessness+leg discomfort | 3 (9) |
| Other | 2 (6) |
Data are presented as mean±sd unless otherwise stated. GOLD: Global Initiative for Chronic Obstructive Lung Disease; SABA: short-acting beta-agonist; SAMA: short-acting muscarinic antagonist; LABA: long-acting beta-agonist; LAMA: long-acting muscarinic antagonist; ICS: inhaled corticosteroid; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; mMRC: modified Medical Research Council; CAT: COPD Assessment Test; HADS: Hospital Anxiety and Depression Scale; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; DLCO: diffusing capacity of the lung for carbon monoxide; R5: resistance at 5 Hz; R20: resistance at 20 Hz; X5: reactance at 5 Hz; V′O2peak: peak oxygen consumption. #: only 20 participants completed the CAT.
31 out of 32 participants completed all 4 min of each 4-min CWR-TT. One participant with COPD completed just 1 min under the F2L condition; their perceptual and physiological responses at iso-time (end of the first minute) were analysed. The average airflow was 3.56±1.57 m·s−1: 17, eight and seven participants self-selected fan settings I (∼2.5 m·s−1), II (∼4.2 m·s−1) and III (∼6.5 m·s−1), respectively. The randomisation sequence and sequence effect analysis results are in supplemental table 1.
As illustrated in figure 1a and table 2, there was a significant intervention main effect on intensity ratings of exertional breathlessness: ratings were lower under F2F than under NF (by 0.5 (95% CI 0.09−0.91) units on the mBorg-CR10), but not different between F2F and F2L, or between F2L and NF. Figure 1d shows the intra-individual changes in exertional breathlessness intensity ratings between conditions: 10 (31.2%), five (15.6%) and 10 (31.2%) out of 32 participants reported a minimal clinically important decrease of ≥1 unit on the mBorg-CR10 scale [24] under F2F versus NF, F2F versus F2L and F2L versus NF, respectively.
FIGURE 1.
Perceptual responses to exercise under no fan (NF), fan-to-leg (F2L) and fan-to-face (F2F) conditions in all participants (n=32): a) breathlessness intensity; b) breathlessness unpleasantness; c) leg discomfort; d) intra-individual changes in exertional breathlessness intensity ratings between experimental conditions. Data are presented as mean±sd. The horizontal red dotted lines represent the minimal clinically important difference (MCID) of ±1 unit on the modified Borg 0–10 Category-Ratio scale (mBorg-CR10). Participants who have missing bars represent the 12, 15 and 14 participants who had no change in their breathlessness intensity under F2F−NF, F2L−NF and F2F−F2L, respectively (panel d). F2F−NF: difference between F2F and NF; F2L−NF: difference between F2L and NF; F2F−F2L: difference between F2F and F2L.
TABLE 2.
Effect of fan-to-face (F2F) therapy on perceptual and physiological responses at the end of the 4-min constant-work rate treadmill exercise test in all participants (n=32)
| Variable | NF | F2L | F2F | Intervention main effect | F2F versus NF | F2L versus NF | F2F versus F2L |
|---|---|---|---|---|---|---|---|
| Breathlessness intensity, mBorg-CR10 | 2.5±1.5 | 2.1±1.5 | 2.0±1.5 | p=0.025 | Mdiff=0.5 (0.09–0.91) | Mdiff=0.4 (−0.01–0.72) | Mdiff=0.1 (−0.21–0.49) |
| η2G=0.020 | p=0.023 | p=0.727 | p=0.134 | ||||
| Hedges’ g=0.329 | Hedges’ g=0.233 | Hedges’ g=0.092 | |||||
| Breathlessness unpleasantness, mBorg-CR10 | 2.4±1.5 | 2.0±1.4 | 2.1±1.4 | p=0.059 | Mdiff=0.3 (−0.05–0.74) | Mdiff=0.4 (−0.00–0.78) | Mdiff=0.0 (−0.23–0.32) |
| η2G=0.014 | Hedges’ g=0.225 | Hedges’ g=0.256 | Hedges’ g=0.032 | ||||
| Leg discomfort, mBorg-CR10 | 2.2±1.9 | 1.8±1.7 | 1.9±1.9 | p=0.060 | Mdiff=0.3 (−0.09–0.63) | Mdiff=0.4 (0.08–0.76) | Mdiff=0.2 (−0.22–0.53) |
| η2G=0.009 | Hedges’ g=0.138 | Hedges’ g=0.233 | Hedges’ g=0.086 | ||||
| Face skin surface temperature, °C | 36.2±0.3 | 36.1±0.5 | 34.6±0.9 | p<0.001 | Mdiff=1.5 (1.26–1.83) | Mdiff=0.1 (−0.10–0.27) | Mdiff=1.5 (1.18–1.73) |
| η2G=0.563 | p<0.001 | p=0.769 | p<0.001 | ||||
| Hedges’ g=2.197 | Hedges’ g=0.192 | Hedges’ g=1.924 | |||||
| Heart rate, bpm | 107.2±16.3 | 102.1±22.0 | 100.9±19.8 | p=0.086 | Mdiff=6.3 (0.47–12.15) | Mdiff=5.1 (0.16–10.12) | Mdiff=1.2 (−6.02–8.36) |
| η2G=0.020 | Hedges’ g=0.339 | Hedges’ g=0.258 | Hedges’ g=0.054 | ||||
| Oxygen saturation, % | 92.3±4.4 | 91.1±4.6 | 90.9±6.2 | p=0.174 | Mdiff=1.4 (−0.45–3.21) | Mdiff=1.2 (−0.22–2.64) | Mdiff=0.2 (−1.41–1.75) |
| η2G=0.015 | Hedges’ g=0.248 | Hedges’ g=0.261 | Hedges’ g=0.031 | ||||
| Respiratory rate, breaths per min | 30.7±7.9 | 30.1±6.7 | 29.8±6.5 | p=0.497 | Mdiff=0.8 (−0.73–2.34) | Mdiff=0.5 (−0.76–1.80) | Mdiff=0.3 (−1.14–1.73) |
| η2G=0.002 | Hedges’ g=0.109 | Hedges’ g=0.069 | Hedges’ g=0.043 | ||||
| Tidal volume, AU | 297.5±121.3 | 303.5±125.6 | 304.6±113.0 | p=0.820 | Mdiff=7.1 (−17.27–31.49) | Mdiff=6.0 (−18.33–30.42) | Mdiff=1.0 (−24.97–27.09) |
| η2G=0.001 | Hedges’ g=0.059 | Hedges’ g=0.048 | Hedges’ g=0.009 | ||||
| Minute ventilation, AU·min−1 | 9.0±4.1 | 8.8±4.0 | 9.1±4.0 | p=0.826 | Mdiff=0.1 (−0.76–0.97) | Mdiff=0.1 (−0.48–0.78) | Mdiff=0.3 (−0.72–1.22) |
| η2G=0.001 | Hedges’ g=0.025 | Hedges’ g=0.03 | Hedges’ g=0.060 |
Data are presented as mean±sd or mean difference (95% CI) unless otherwise stated. The intervention main effect p-value and partial η2 (η2G) represent the results of the comparison between the three conditions. NF: no fan; F2L: fan to leg; mBorg-CR10: modified Borg 0–10 Category-Ratio scale; Mdiff: mean difference; bpm: beats per min; AU: arbitrary units.
Tface was ∼1.5 °C lower during exercise under F2F than under NF or F2L; however, there was no intervention main effect on breathlessness unpleasantness rating, leg discomfort rating, HR, SpO2, RR, VT or V′E during exercise (figure 2, table 2).
FIGURE 2.
Physiological responses to exercise under no fan (NF), fan-to-leg (F2L) and fan-to-face (F2F) conditions in all participants (n=32). Data are presented as mean±sd. Participants who could not wear the Hexoskin Smart Shirt™ due to sizing issues (n=2) or when its data could not be analysed due to software issues (n=3) were excluded from analysis. SpO2: peripheral oxygen saturation; AU: arbitrary units.
To reduce potential bias from the imbalance of participants with COPD (n=27) and those with ILD (n=5), we compared exertional breathlessness intensity ratings between F2F, F2L and NF conditions in COPD participants only. There was a significant main intervention effect (p=0.035), with a significant difference in breathlessness intensity ratings of 0.6 (95% CI 0.07−1.04) units on the mBorg-CR10 between F2F and NF, with no difference between F2F and F2L, or between F2L and NF (supplemental figure 3a, supplemental table 2).
Subgroup analysis
Participants in the exploratory (unplanned) subgroup analysis were symptomatic adults with COPD (n=16, 80%) or ILD (n=4, 20%), characterised by impaired lung function and exercise intolerance mostly due to intolerable breathlessness (supplemental table 3). The average airflow was 3.93±1.68 m·s−1: 10, five and five participants self-selected fan settings I (∼2.5 m·s−1), II (∼4.2 m·s−1) and III (∼6.5 m·s−1), respectively. The randomisation sequence and sequence effect analysis results are in supplemental table 4.
There was a significant intervention main effect on intensity ratings of exertional breathlessness (figure 3a, supplemental table 5): ratings were lower under F2F than under NF (by 0.9 (95% CI 0.32−1.43) units on the mBorg-CR10), but not different between F2F and F2L, or between F2L and NF. Nine (45%), five (25%) and eight (40%) out of 20 participants reported a minimal clinically important decrease of ≥1 unit on the mBorg-CR10 scale [24] under F2F versus NF, F2F versus F2L and F2L versus NF conditions, respectively (figure 3d).
FIGURE 3.
Perceptual responses to exercise under no fan (NF), fan-to-leg (F2L) and fan-to-face (F2L) conditions in the exploratory subgroup of 20 participants who reported at least moderately severe breathlessness during exercise under the NF condition: a) breathlessness intensity; b) breathlessness unpleasantness; c) leg discomfort; d) intra-individual changes in exertional breathlessness intensity ratings between experimental conditions. Data are presented as mean±sd. The horizontal red dotted lines represent the minimal clinically important difference (MCID) of ±1 unit on the modified Borg 0–10 Category-Ratio scale (mBorg-CR10). Participants who have missing bars represent the eight, nine and nine participants who had no change in their breathlessness intensity under F2F−NF, F2L−NF and F2F−F2L, respectively (panel d). F2F−NF: difference between F2F and NF; F2L−NF: difference between F2L and NF; F2F−F2L: difference between F2F and F2L.
There was a significant intervention main effect on exertional breathlessness unpleasantness ratings and Tface (figure 3b, supplemental table 5): breathlessness unpleasantness ratings were lower under both F2F (by 0.7 (95% CI 0.18−1.17) units on the mBorg-CR10) and F2L (by 0.6 (95% CI 0.03−1.12) units on the mBorg-CR10) than under NF, but F2F and F2L were not different from each other. Tface was ∼1.5 °C lower under F2F than under NF or F2L, which again were not different from each other. There was no intervention main effect on intensity ratings of leg discomfort, HR, SpO2, RR, VT or V′E during exercise (figure 4, supplemental table 5).
FIGURE 4.
Physiological responses to exercise under no fan (NF), fan-to-leg (F2L) and fan-to-face (F2F) conditions in the exploratory subgroup of 20 participants who reported at least moderately severe breathlessness during exercise under the NF condition. Data are presented as mean±sd. SpO2: peripheral oxygen saturation; AU: arbitrary units.
Among the COPD-only participants in the exploratory subgroup analysis (n=16), there was a significant main intervention effect (p=0.009), with significantly lower breathlessness intensity ratings during F2F, by 1.0 (95% CI 0.32−1.68) unit on the mBorg-CR10, than under NF, and with no difference between F2F and F2L, or between F2L and NF (supplemental figure 3b, supplemental table 2).
Discussion
The main finding of this randomised controlled crossover trial is that F2F therapy reduced exertional breathlessness intensity compared with NF (but not F2L) in symptomatic adults with chronic respiratory disease. This benefit of F2F versus NF was particularly notable among those with more severe exertional breathlessness, though again not greater than F2L. In both the main and the exploratory subgroup analyses, reductions in breathlessness under F2F versus NF occurred alongside decreases in Tface, but without contemporaneous changes in HR, SpO2, RR, VT or V′E.
We studied older adults with COPD or ILD who, despite receiving guideline-based treatment(s), experienced disabling breathlessness and had impaired exercise tolerance. These individuals represent a clinical population in need of symptom-specific adjuvant interventions. Across all participants, F2F produced a modest yet significant reduction of 0.5 units on the mBorg-CR10 scale in exertional breathlessness intensity versus NF, with no difference between F2F and F2L or between F2L and NF. Approximately 30% of participants reported a clinically important decrease (≥1 unit) under both F2F and F2L versus NF. Among those with at least moderate exertional breathlessness under NF in the exploratory subgroup analysis, breathlessness intensity and unpleasantness ratings were 0.9 and 0.7 units lower, respectively, under F2F than under NF. No differences in exertional breathlessness were found between F2F and F2L in this subgroup (possibly reflecting a lack of statistical power), and F2L unexpectedly lowered breathlessness unpleasantness compared with NF. The modest end-exercise breathlessness intensity ratings in the overall sample highlight the challenge of eliciting symptoms of sufficient intensity to be amenable to intervention during standardised exercise testing in people with chronic lung disease, as previously described by our research group for use of F2F therapy during exercise training of people with chronic lung disease [9]. Relief of exertional breathlessness with F2F versus NF occurred without changes in cardiorespiratory variables (e.g., V′E), consistent with previous studies [4, 15, 16].The optimal airflow speed of 3.56±1.57 m·s−1 selected by our participants was lower than the 4.27 m·s−1 used by Marchetti et al. [4], higher than the 1.5 m·s−1 and 2.85 m·s−1 used by Venkatapen et al. [15] and Luckett et al. [8], respectively, but closer to recommendations by Brew et al. [27], suggesting a potential sweet spot (i.e., “Goldilocks airflow”) for optimal comfort and symptom relief.
A key limitation of F2F studies is the inability to blind participants to the F2F intervention, introducing the possibility of a placebo effect despite concealing the study hypothesis. Unlike prior studies that used a single comparator (F2L or NF) among 10–25 people with chronic lung disease [4, 15, 16], we included both F2L and NF controls in an adequately powered sample. Still, the 0.5-unit reduction in exertional breathlessness intensity between F2F and NF was modest, and there was no difference between F2F and F2L, supporting a potential placebo effect. While the exploratory subgroup analyses revealed a larger effect (0.9 units) on intensity ratings of exertional breathlessness for F2F versus NF, F2F and F2L remained comparable, preventing definitive conclusion on the mechanism of action. Thus, while our findings may reflect central neuromodulation of exertional breathlessness via activation of TRPM8 receptors on the trigeminal nerve [10, 11], we cannot rule out a placebo effect that may have been caused by non-trigeminal-specific airflow effects. Generalisability is also limited by the controlled laboratory setting; real-world use of F2F therapy involves diverse fan types, varied techniques and triggers beyond exercise (e.g., anxiety, air pollution). Future research should focus on the effectiveness of F2F therapy in daily life contexts and in response to the multiple different causes of episodic breathlessness. Indeed, we recently reported that adding F2F therapy to a supervised exercise training programme may allow people with chronic lung disease to train at their prescribed or even higher than prescribed exercise training volumes with lower exertional symptoms and fewer subjective perceptions of respiratory discomfort, underscoring the utility of F2F in various contexts [9]. Furthermore, a range of preferred airflow speeds were self-selected by our participants, providing unique insight into the subjectivity of how optimal airflow speed is perceived, and possibly contributing to the modest overall effect. However, our exploratory subgroup and COPD-only analyses support our results in the main analysis, which is that F2F alleviates breathlessness compared with NF. Heterogeneity in participant characteristics and variability in the exercise intensity of the 4-min CWR-TT, given limited data on its repeatability and responsiveness, may also have influenced our results. Finally, while breathlessness is common in both COPD and ILD, future research should include larger samples of people with ILD to better evaluate disease-specific responses to F2F therapy.
Research implications
The decrease in exertional breathlessness with F2F therapy was most notable in our subgroup of 20 participants reporting at least moderate exertional breathlessness under the NF condition. This highlights the importance of ensuring that the interventional target (breathlessness) is present to a degree that is amenable to change (i.e., even a sharpshooter can't hit a target that doesn't exist); otherwise, efficacy may be underestimated. Although an algorithm exists to elicit a target intensity of exertional breathlessness [28], and we used an exertional breathlessness intensity threshold of ≥3 units on the mBorg-CR10 scale for our exploratory subgroup analysis based on our previous work [29], no consensus exists on the optimal breathlessness intensity threshold for evaluating interventional efficacy. Future studies should aim to define the minimum exertional breathlessness intensity to assess treatment effects, and recruit sample sizes that meet or exceed this threshold. Further work should also determine whether baseline characteristics (e.g., modified Medical Research Council dyspnoea rating, COPD Assessment Test score) predict responsiveness to F2F therapy.
Clinical implications and conclusions
Our findings suggest that F2F therapy may alleviate exertional breathlessness in people with chronic respiratory disease, particularly those with more severe exertional breathlessness. Although the effect of F2F therapy on exertional breathlessness appears modest and may partially reflect a placebo effect, given no differences between F2F and F2L, the observed 1.5 °C reduction in Tface, alongside unchanged objective cardiorespiratory measures such as V′E, supports a possible central neuromodulatory mechanism involving stimulation of intranasal/upper-airway TRPM8 receptors on the trigeminal nerve. This aligns with work by Schaeffer et al. [30], who reported a modest yet significant benefit of inhaled menthol (versus placebo) on exertional breathlessness in people with COPD; menthol is thought to have a similar TRPM8 receptor-mediated central mechanism of action [31].
Consistent with extant literature [32], F2F therapy was well tolerated with no adverse events, and participants self-selected a subjective “optimal” facial airflow of ∼3.56 m·s−1. Whether relief reflects placebo and/or central neuromodulation of respiratory symptom perception, F2F (compared with NF) was associated with clinically meaningful relief of exertional breathlessness in ∼30% of our participants overall and ∼45% of those with more severe exertional breathlessness. In people with COPD only (to minimise the bias caused by the imbalance of COPD and ILD in the study sample), the effect of F2F versus NF remained the same. Given its simplicity, absence of harms, feasibility, low cost, acceptability [16, 33] and accessibility, F2F therapy represents a potentially effective nonpharmacological intervention that healthcare providers should reasonably consider recommending to symptomatic adults with chronic respiratory disease to help manage their exertional breathlessness at home or in the clinic care setting.
Acknowledgements
The authors thank the participants for their time and cooperation; the multidisciplinary clinical support staff from the Montreal Chest Institute Pulmonary Rehabilitation Program, Nathalie Saad and Yujie Hu from the Jewish General Hospital, Nicole Ezer from the McGill University Health Centre for their continued support and assistance in participant recruitment; and Justin Riffou, Erica Delucilla, Aleksander Ilic and Anna Blinova for their help with data acquisition. The authors would like to thank the Quebec Respiratory Health Research Network for its financial support of this project.
Footnotes
Provenance: Submitted article, peer reviewed.
This clinical trial is prospectively registered with ClinicalTrials.gov as NCT05659550.
Ethics statement: This randomised controlled crossover trial received ethics approval from the Research Institute of the McGill University Health Centre (IRB# 2023-9286).
Author contributions: Study conception and design: R. Aucoin, H. Lewthwaite and D. Jensen. Data acquisition: R. Aucoin and D. Nguyen. Data analysis: R. Aucoin, D. Nguyen and D. Jensen. Manuscript writing: R. Aucoin and D. Jensen. Manuscript editing: R. Aucoin, H. Lewthwaite, M. Ekström, A. von Leupoldt and D. Jensen. Manuscript approval prior to submission: R. Aucoin, D. Nguyen, B. Ross, J. Bourbeau, H. Lewthwaite, M. Ekström, A. von Leupoldt and D. Jensen.
Conflict of interest: D. Jensen holds a Canada Research Chair in Clinical Exercise and Respiratory Physiology (Tier 2) from the Canadian Institutes of Health Research. No conflicts of interest relevant to this article, financial or otherwise, are declared by the authors.
Support statement: This work was supported by an operating grant to D. Jensen from the Quebec Respiratory Health Research Network paid to McGill University. D. Jensen holds a Canada Research Chair in Clinical Exercise and Respiratory Physiology (Tier 2) from the Canadian Institutes of Health Research. R. Aucoin was supported by doctoral fellowships from the Fonds de Recherche du Québec (Santé) and the Canadian Respiratory Research Network. A. von Leupoldt was supported by grants from the Research Foundation Flanders (FWO), Belgium (G004624N, G0C1921N), by an infrastructure grant from the FWO and the Research Fund KU Leuven, Belgium (I011320N; AKUL/19/06) and by a project grant of the Research Fund KU Leuven, Belgium (C16/23/002). M. Ekström was supported by an unrestricted grant from the Swedish Research Council (Dnr: 2019–02081).
Supplementary material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01636-2025.SUPPLEMENT
Data availability
The data can be viewed by contacting the corresponding author.
References
- 1.Parshall MB, Schwartzstein RM, Adams L, et al. An official American Thoracic Society statement: update on the mechanisms, assessment, and management of dyspnea. Am J Respir Crit Care Med 2012; 185: 435–452. doi: 10.1164/rccm.201111-2042ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Müllerová H, Lu C, Li H, et al. Prevalence and burden of breathlessness in patients with chronic obstructive pulmonary disease managed in primary care. PLoS ONE 2014; 9: e85540. doi: 10.1371/journal.pone.0085540 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.van den Bosch L, Wang T, Bakal JA, et al. A retrospective, descriptive study of dyspnea management in a multidisciplinary interstitial lung disease clinic. Am J Hosp Palliat Care 2023; 40: 153–163. doi: 10.1177/10499091221096416 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Marchetti N, Lammi MR, Travaline JM, et al. Air current applied to the face improves exercise performance in patients with COPD. Lung 2015; 193: 725–731. doi: 10.1007/s00408-015-9780-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Swan F, Newey A, Bland M, et al. Airflow relieves chronic breathlessness in people with advanced disease: an exploratory systematic review and meta-analyses. Palliat Med 2019; 33: 618–633. doi: 10.1177/0269216319835393 [DOI] [PubMed] [Google Scholar]
- 6.Kako J, Morita T, Yamaguchi T, et al. Fan therapy is effective in relieving dyspnea in patients with terminally ill cancer: a parallel-arm, randomized controlled trial. J Pain Symptom Manage 2018; 56: 493–500. doi: 10.1016/j.jpainsymman.2018.07.001 [DOI] [PubMed] [Google Scholar]
- 7.Yu S, Sun K, Xing X, et al. Fan therapy for the relief of dyspnea in adults with advanced disease and terminal illness: a meta-analysis of randomized controlled trials. J Palliat Med 2019; 22: 1603–1609. doi: 10.1089/jpm.2019.0140 [DOI] [PubMed] [Google Scholar]
- 8.Luckett T, Roberts M, Swan F. Fan therapy for breathlessness – how do you do it and why? Curr Opin Support Palliat Care 2025; 9: 111–116. doi: 10.1097/SPC.0000000000000752 [DOI] [PubMed] [Google Scholar]
- 9.Aucoin R, Nguyen D, Ross B, et al. Facial airflow enhances the benefits of exercise training in people with chronic lung disease: a randomized controlled trial. Eur Respir J 2025: 67: 2501109. doi: 10.1183/13993003.01109-2025 [DOI] [PubMed] [Google Scholar]
- 10.Aucoin R, Lewthwaite H, Ekström M, et al. Impact of trigeminal and/or olfactory nerve stimulation on measures of inspiratory neural drive: implications for breathlessness. Respir Physiol Neurobiol 2023; 311: 104035. doi: 10.1016/j.resp.2023.104035 [DOI] [PubMed] [Google Scholar]
- 11.Aucoin R, Lewthwaite H, Ekström M, et al. Impact of trigeminal nerve and/or olfactory nerve stimulation on activity of human brain regions involved in the perception of breathlessness. Respir Physiol Neurobiol 2023; 311: 104036. doi: 10.1016/j.resp.2023.104036 [DOI] [PubMed] [Google Scholar]
- 12.Tsai JJ, Chen KH, Fang HF, et al. Relieving from breathlessness in the wind: a meta-analysis and subjective report of effectiveness of fan blowing in patients with cardiorespiratory diseases or cancer. Am J Hosp Palliat Care 2022; 39: 977–985. doi: 10.1177/10499091211056327 [DOI] [PubMed] [Google Scholar]
- 13.Da Purificaçao J, Pierret C, Bonnevie T, et al. Acute effects of fan therapy on exercise-induced breathlessness in healthy adults and people living with chronic respiratory disease: a systematic review. Respir Med Res 2025; 88: 101172. doi: 10.1016/j.resmer.2025.101172 [DOI] [PubMed] [Google Scholar]
- 14.Leivo-Korpela S, Rantala HA, Lehtimäki L, et al. Management of dyspnea with high-flow nasal air or fan – a randomized controlled crossover trial. J Pain Symptom Manage 2026; 71: 8–15. doi: 10.1016/j.jpainsymman.2025.08.033 [DOI] [PubMed] [Google Scholar]
- 15.Venkatapen D, Caty G, Froidure A, et al. Effect of fan therapy on dyspnea in patients with chronic respiratory diseases during exercise: a randomized controlled trial. Respir Med 2025; 249: 108426. doi: 10.1016/j.rmed.2025.108426 [DOI] [PubMed] [Google Scholar]
- 16.Long A, Cartwright M, Reilly CC. Impact of fan therapy during exercise on breathlessness and recovery time in patients with COPD: a pilot randomised controlled crossover trial. ERJ Open Res 2021; 7: 00211. doi: 10.1183/23120541.00211-2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.von Leupoldt A, Karachi C, Jelinčić V. Relieving dyspnoea through the brain. Eur Respir J 2024; 64: 2401828. doi: 10.1183/13993003.01828-2024 [DOI] [PubMed] [Google Scholar]
- 18.Bestall JC, Paul EA, Garrod R, et al. Usefulness of the Medical Research Council (MRC) Dyspnoea Scale as a measure of disability in patients with chronic obstructive pulmonary disease. Thorax 1999; 54: 581–586. doi: 10.1136/thx.54.7.581 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Nemeş R, Bumbăcea D. [Ways to quantify dyspnea severity. Part I. Baseline Dyspnea Index (BDI), Transition Dyspnea Index (TDI)]. Pneumologia 2006; 55: 58–63. [PubMed] [Google Scholar]
- 20.Jones PW, Harding G, Berry P, et al. Development and first validation of the COPD assessment test. Eur Respir J 2009; 34: 648–654. doi: 10.1183/09031936.00102509 [DOI] [PubMed] [Google Scholar]
- 21.Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361–370. doi: 10.1111/j.1600-0447.1983.tb09716.x [DOI] [PubMed] [Google Scholar]
- 22.Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982; 14: 377–381. [PubMed] [Google Scholar]
- 23.Lewthwaite H, Jensen D, Ekström M. How to assess breathlessness in chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis 2021; 16: 1581–1598. doi: 10.2147/COPD.S277523 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ries AL. Minimally clinically important difference for the UCSD shortness of breath questionnaire, Borg scale, and visual analog scale. COPD 2005; 2: 105–110. doi: 10.1081/COPD-200050655 [DOI] [PubMed] [Google Scholar]
- 25.Abdallah SJ, Wilkinson-Maitland C, Saad N, et al. Effect of morphine on breathlessness and exercise endurance in advanced COPD: a randomised crossover trial. Eur Respir J 2017; 50: 1701235. doi: 10.1183/13993003.01235-2017 [DOI] [PubMed] [Google Scholar]
- 26.Borel B, Wilkinson-Maitland CA, Hamilton A, et al. Three-minute constant rate step test for detecting exertional dyspnea relief after bronchodilation in COPD. Int J Chron Obstruct Pulmon Dis 2016; 11: 2991–3000. doi: 10.2147/COPD.S113113 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Brew A, O'Beirne S, Johnson MJ, et al. Airflow rates and breathlessness recovery from submaximal exercise in healthy adults: prospective, randomised, cross-over study. BMJ Support Palliat Care 2024; 14: 442–448. doi: 10.1136/spcare-2023-004309 [DOI] [PubMed] [Google Scholar]
- 28.Maltais F, Aumann JL, Kirsten AM, et al. Dual bronchodilation with tiotropium/olodaterol further reduces activity-related breathlessness versus tiotropium alone in COPD. Eur Respir J 2019; 53: 1802049. doi: 10.1183/13993003.02049-2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tracey L, Murray S, Bui KL,et al. 3-minute constant-rate shuttle walking (3MWT) and stair stepping tests (3MST) to assess exertional breathlessness in COPD: responsiveness to pulmonary rehabilitation and estimation of the minimal important difference. Eur Respir J 2019; 54: suppl 63, PA511. doi: 10.1183/13993003.congress-2019.PA511 [DOI] [Google Scholar]
- 30.Schaeffer MR, Vanden Bossche L, Beckers K, et al. Inhaled menthol for dyspnea relief during cycle exercise in COPD: a randomized trial. Chest 2025; 168: 390–401. doi: 10.1016/j.chest.2025.03.002 [DOI] [PubMed] [Google Scholar]
- 31.Bossche LV, Piers S, Schaeffer MR, et al. Menthol inhalation relieves dyspnoea through the brain. ERJ Open Res 2026; in press ( 10.1183/23120541.01164-2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Luckett T, Phillips J, Johnson MJ, et al. Contributions of a hand-held fan to self-management of chronic breathlessness. Eur Respir J 2017; 50: 1700262. doi: 10.1183/13993003.00262-2017 [DOI] [PubMed] [Google Scholar]
- 33.Khor YH, Saravanan K, Holland AE, et al. A mixed-methods pilot study of handheld fan for breathlessness in interstitial lung disease. Sci Rep 2021; 11: 6874. doi: 10.1038/s41598-021-86326-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01636-2025.SUPPLEMENT
Data Availability Statement
The data can be viewed by contacting the corresponding author.